Background Classical integer-order membrane and bioelectrical circuit models are useful for idealized ion-channel behavior, but they often fail to reproduce the distributed relaxation, power-law impedance, depressed Nyquist arcs, and memory-dependent phase dispersion observed in biological membranes and tissue interfaces. Objectives This study aims to develop a fractional four-element Windkessel-constant phase framework, denoted BioMem-FWK4α, for analog simulation and frequency-domain interpretation of biological ion-channel and membrane-like impedance responses. Design A theoretical, computational, and analog-circuit modeling study was conducted using normalized fractional circuit parameters. The study is intended as a mechanistic modeling framework rather than as a direct experimental validation study. Methods The classical four-element Windkessel topology was reformulated by replacing ideal compliance with a constant-phase element of order α and by introducing fractional inertial behavior of order β and an active/passive resistance ratio γ. Caputo and Caputo-Fabrizio formulations were compared under explicit zero-initial frequency-domain assumptions. The model was analyzed through transfer functions, Bode plots, Nyquist loci, Nichols diagrams, RC-ladder approximation, sensitivity indicators, and biological interpretation tables. Results The model predicts non-integer gain slopes approaching -20α dB per decade, phase plateaus near -90α degrees, and depressed Nyquist arcs whose depression increases as α decreases. The γ parameter controls the transition from passive leakage-dominated filtering to active or mixed active-passive regimes, while β modifies delayed ionic or kinetic effects. The RC-ladder approximation supports the physical realizability of the fractional element, although biological validation requires future fitting against patch-clamp and bioimpedance datasets. Conclusion BioMem-FWK4α provides a compact and interpretable circuit framework for describing memory-rich membrane impedance and ion-channel-like dynamics. It does not replace Hodgkin-Huxley or Markov-state models for detailed nonlinear gating, but complements them by offering a reduced-order frequency-domain and analog-realizable representation of distributed membrane memory. Future work should validate the model using time-resolved patch-clamp, tissue bioimpedance, and voltage-dependent nonlinear parameter identification.
Background Automated facial wrinkle detection is relevant to facial-ageing assessment, cosmetic analysis, dermatological screening, and personalized skincare. However, wrinkles are thin, low-contrast, and spatially irregular structures whose appearance may be affected by illumination, pose, skin texture, age, image quality, and demographic characteristics. Objective This study proposes FWD-DHRN-HSWOA, a facial wrinkle-detection framework combining attentional image enhancement, high-resolution feature representation, and metaheuristic parameter optimization. Methods Facial images were obtained from the FG-NET Aging Database and supplemented with study-specific wrinkle annotations because wrinkle labels are not native to FG-NET. Images were aligned, resized, normalized, and enhanced using Deep Attentional Guided Image Filtering. A Dynamic Lightweight High-Resolution Network was then used to preserve fine spatial information during wrinkle localization. Harbor Seal Whiskers Optimization was applied as an outer-loop procedure for tuning selected model parameters. The proposed method and implemented baselines were evaluated using the same data partitions, preprocessing conditions, and performance metrics. Results Within the evaluated FG-NET-derived annotation setting, the proposed framework produced higher observed accuracy, precision, recall, F1-score, and specificity and lower RMSE than the implemented comparison models. These results reflect performance under the reported experimental setting and should not be interpreted as evidence of equivalent performance across unrepresented demographic groups or dedicated clinical wrinkle datasets. Conclusion Combining attentional preprocessing, high-resolution feature extraction, and metaheuristic optimization shows promise for fine facial wrinkle localization. Nevertheless, external validation on dedicated wrinkle datasets with verified age, ethnicity, skin-tone, and acquisition-condition diversity is necessary before broad deployment.
Background Diagnosis of elbow instability in clinical practice commonly relies on manual stress testing by the physician with or without combination of fluoroscopy or ultrasound, which may result in variable force application and subjective interpretation. A standardized method for applying controlled mechanical loading during dynamic imaging may improve reproducibility. Objective The objective of this study was to develop and perform initial cadaveric feasibility testing of a CT-compatible diagnostic device for device-guided assessment of elbow instability with simultaneous force and positional measurement. Design Technical development study with initial cadaveric feasibility testing. Methods Building upon the initial prototype the present paper, describes the development of MELBO the follow-up prototype with adjusted kinematics and improved stability, the validation of the force sensor is addressed, and the first cadaver test results are presented. Results The MELBO diagnostic tool enabled a controlled device-guided elbow motion during 4D-CT acquisition and allowed simultaneous recording of force and positional data. The acquired dynamic CT datasets were suitable for qualitative assessment of elbow joint stability. No relevant device-related imaging artifacts were observed. Conclusion The present study demonstrates the technical feasibility of device-guided simultaneous force and positional measurement during 4D-CT using the MELBO diagnostic tool. These findings provide the basis for further cadaveric studies and future clinical investigations aimed at quantitative assessment of elbow instability under standardized loading conditions.
Objectives We developed and evaluated a lightweight, interpretable, and computationally efficient hybrid deep learning model for multiclass classification of early lung cancer in low-dose CT scans, potentially deployable in resource-limited healthcare environments. Methods A novel hybrid architecture was developed that integrates Bidirectional Long Short-Term Memory (Bi-LSTM) networks, Temporal Convolutional Networks (TCNs), Efficient Channel Attention (ECA) blocks, and Local Interpretable Model-Agnostic Explanations (LIME). The model employed depthwise separable convolutions to reduce computational complexity. A multi-stream feature extraction framework was implemented to enhance interpretability and capture spatial-temporal patterns. The model was trained and validated on the IQ-OTH/NCCD dataset (version 2), containing 3,609 CT scan slices from 110 patients (1,097 original, 2,512 augmented), across three classes: normal, benign, and malignant. The dataset was split into training (60%), validation (30%), and testing (10%) subsets. Training was conducted using the AdamW optimizer for 16 epochs. Results The model achieved 98.06% accuracy, 98.15% precision, 98.06% recall, and 98.04% F1-score, with a 99.88% AUC, a model size of 3.33 MB, and 279,561 parameters. Conclusion The lightweight model achieves high diagnostic accuracy with computational efficiency, SHAP-based and LIME-based interpretability methods, enabling potential suitability for deployment in resource-constrained clinical settings.
Background Acute Myelogenous Leukemia (AML) is a rapidly progressing blood and bone marrow cancer, prevalent among adults. Very low five-year survival rate, non-specific and ambiguous symptoms, and lack of potent screenings make early detection and prompt treatment crucial, especially for younger patients. The need for multiple tests to confirm AML and possible misdiagnosis often leads to multiple consultations before moving to the next stage of investigations. This can create operational bottlenecks in the hematology department, leading to further delays in the diagnostic pathway. Objective To develop a lightweight artificial intelligence (AI) model using complete blood count (CBC) data for early AML detection and development of a decision support system (DSS) for triage support. Methods The data for this study were retrieved retrospectively from the National Center for Cancer Care and Research (NCCCR), Qatar (2016-2022). We used 510 CBC data records of AML and non-AML individuals. Statistical analysis of CBC features (mean ± SD) for AML and control patients was performed, followed by principal component analysis (PCA) to assess the predictive ability of CBC alone. Results Rigorous feature selection and model tuning resulted in a predictive diagnostic Support Vector Machine (SVM) model to alleviate delays in the AML care pathway. Employing a five-fold cross-validation approach, achieved an accuracy of 96.4% (S.D. 0.029) for test set and 80% (S.D. 0.036) for validation set. This 80% accuracy on the validation set, with high sensitivity (100% recall) but lower precision (77.1%), represents an acceptable triage trade-off, although it may increase follow-up referrals and workload. The developed model showed encouraging performance when used to detect AML using CBCs taken up to 1-year prior diagnosis. Conclusion This study emphasizes that, with routine CBC data, we can enable better patient screening and referral in the initial stages of triage through a cost-effective, AI-based decision support system that provides complementary support to doctors for timely AML detection.
The carotid bifurcation plays a crucial role in cerebral perfusion, and its hemodynamic behavior is influenced by external factors, including interactions with surrounding anatomical structures. This study investigates the impact of hyoid bone-induced compression on carotid artery hemodynamics using computational fluid dynamics (CFD) and fluid-structure interaction (FSI) modeling. The results reveal significant alterations in time-averaged wall shear stress (TAWSS), oscillatory shear index (OSI), velocity distribution, and Von-Mises stress, highlighting the biomechanical effects of external compression at various arterial locations. Compression of the common carotid artery results in a marked reduction in downstream TAWSS, dropping from 3.68 Pa to 1.70 Pa, and a substantial increase in OSI, reaching up to 0.46, suggesting disturbed flow patterns that may contribute to pathological vessel remodeling. In contrast, compression at the internal carotid artery leads to localized elevations in TAWSS, reaching approximately 6.90 Pa at the contact site and 7.80 Pa at the bifurcation, while OSI remains relatively low (around 0.22). Similarly, high oscillatory low magnitude shear index (HOLMES) levels decrease significantly after compression of the common carotid artery, from 1.23 Pa to 0.41 Pa, but increase when the internal carotid artery is compressed, peaking at 3.30 Pa at the contact site and 2.60 Pa downstream within the external carotid artery. Velocity streamline analysis demonstrates prominent vortex formation at the bifurcation, particularly in cases where direct compression occurs at this site, indicating disrupted and recirculating flow. Von-Mises stress analysis shows the highest stress concentration at the contact region across all cases. The lowest stress is observed with common carotid compression (approximately 0.6 MPa), while higher stresses occur in the bifurcation and internal carotid cases (around 1.5 MPa). The greatest mechanical stress, reaching approximately 1.9 MPa, is seen when the external carotid artery is compressed, indicating a higher risk of structural damage in this region. These results improve understanding of the effects of external anatomical interactions on carotid artery hemodynamics and motivate further investigation of their clinical significance. Future investigations should prioritize patient-specific modeling and in vivo validation to better assess the long-term vascular health impacts of hyoid bone-induced compression.
Background T-wave alternans (TWA) refers to variations in the ventricular repolarization pattern observed on the ECG, which has been associated with cardiac instability and an increased risk of sudden cardiac death. Recently, machine learning (ML) methods have been developed for TWA detection, but their black-box nature limits interpretability. Objectives To address this gap, we propose manifold learning (MnL) to enhance the explainability of these learning models while maintaining TWA detection effectiveness. Methods We fine-tuned nonlinear dimension reduction techniques such as Uniform Manifold Approximation and Projection (UMAP), Isometric Mapping (Isomap), and autoencoders (AE) in combination with ML methods, namely K-nearest neighbors (KNN), random forest (RF), and neural networks (NN). Performance was evaluated using mean and standard deviation across patient-wise permutations. Results In the design stage, the AE-based NN effectively retained essential discriminative information (F1-score 92.1 ± 2.4 %). MnL-generated spaces consistently revealed that misclassifications primarily lie close to the decision boundary and are predominantly associated with lower TWA voltages, which are more dispersed within the space. For ambulatory TWA detection, Isomap combined with RF and the AE-based NN achieved performance comparable to using the complete set of features derived from established TWA analysis methods (F1-score 78.5 ± 6.4 % and 77.9 ± 5.4 %, respectively), including spectral, time-domain, and correlation-based descriptors. The latent space visualization shows that predictions that ultimately become detections are located farther away from the decision boundary. Conclusion MnL-generated spaces provide valuable insights into how classification models differentiate between TWA and non-TWA instances, as well as the patterns in TWA event amplitudes. This approach helps bridge the gap between performance and transparency, supporting more clinically reliable TWA detection.
Background: Pupil size variability (PSV) has emerged as a potential non-contact indicator of autonomic nervous system (ANS) function; however, its physiological origins remain unclear. This study aims to develop a computational model to investigate the physiological foundations of PSV and assess how frequency-domain indices reflect cardiovascular autonomic balance. Methods: We integrated a well-established cardiovascular regulation model with a biomechanical pupillary muscle plant. The model simulates PSV alongside heart rate variability (HRV) by transmitting respiratory and baroreflex inputs through an indirect neural pathway to the pupillary muscles. Frequency-domain analyses were conducted to compare simulated PSV and HRV across different autonomic states. Results: Simulations suggest that PSV arises from respiratory and baroreceptor inputs, with its classical range-nonlinearity (RNL) property emerging naturally from iris biomechanics. The model reproduces key physiological behaviors, including inspiration-linked dilation and parasympathetic modulation. Frequency-domain analyses reveal low- and high-frequency components in PSV that are similar to those found in HRV. However, the magnitudes of these PSV components depend heavily on the underlying autonomic state and the mean pupil size. Conclusion: These findings provide a mechanistic framework for interpreting PSV as a cardiovascular autonomic biomarker. This ultimately supports the development of non-invasive, wearable, and eye-based ANS monitoring systems.
Objectives Artificial intelligence (AI)-driven automated crown design is rapidly transforming digital restorative dentistry by enabling anatomically precise and functionally integrated crowns. This systematic review and meta-analysis critically evaluate whether AI-assisted crown design systems, including machine learning (ML), deep learning (DL), generative adversarial networks (GANs), and diffusion models, produce restorations with comparable or superior morphological accuracy, occlusal integration, internal fit, and workflow efficiency relative to computer-aided design (CAD) or technician-driven workflows. Methods A comprehensive search of MEDLINE (PubMed), Scopus, Web of Science, Embase, and Cochrane Library was conducted for studies published through 17 March 2026. Eligible studies included in vitro, in silico, and clinical investigations comparing AI-based crown design systems with conventional workflows. Primary outcomes were morphological accuracy root-mean-square (RMS) deviation, cusp morphology, volumetric/linear deviation, occlusal contact fidelity, and internal fit; secondary outcomes included marginal adaptation and restoration design time. Risk of bias was assessed using validated tools, and meta-analyses were conducted using random-effects models with standardized mean differences (SMDs). Results Seventeen studies met the inclusion criteria, of which 13 were included in the quantitative synthesis. AI-based systems achieved clinically acceptable morphological accuracy, internal fit, and occlusal contact reproduction (RMS deviation: SMD = −0.15, 95% CI −1.04 to 0.74). Workflow efficiency improved significantly, with reductions in design time of 25-50% and enhanced precision in chamfer and marginal gaps (p < 0.001). DL and GAN-based platforms consistently produced crowns within clinically acceptable deviation ranges (<100–200 μm). Integration of patient-specific occlusal and mandibular dynamics further enhanced functional occlusal prediction. Expert technician refinement remained beneficial in anatomically complex cases. Conclusions AI-assisted crown design demonstrates promising potential for providing reproducible, morphologically accurate, and functionally integrated restorations while potentially enhancing workflow efficiency. This review underscores the potential of AI systems to standardize restorative outcomes and reduce operator dependency, while combined human-AI workflows may enhance performance in complex cases. However, the current evidence is derived predominantly from in vitro and computational studies, with limited prospective clinical validation, limited integration of patient-specific dynamic occlusal data, and insufficient long-term follow-up evidence. Therefore, the findings should be interpreted cautiously and not considered definitive evidence of clinical superiority over conventional workflows. Standardized clinical protocols and prospective trials are required to confirm long-term efficacy.
Background fractional-order modeling provides a powerful framework for representing memory-dependent conduction in excitable biological media. However, existing soliton-based models of myelinated nerve fibers are often theoretical, operator-specific, and insufficiently benchmarked in terms of numerical reproducibility, physiological plausibility, and computational cost. Objectives This study aims to compare the Liouville-Caputo, Atangana-Baleanu, and Beta fractional operators for modeling soliton-like action-potential propagation in ephaptically coupled myelinated nerve fibers, with emphasis on waveform stability, energy retention, biological consistency, computational efficiency, and adaptive parameter learning. Design A comparative computational modeling study was conducted using a coupled fractional nonlinear partial differential equation framework, physiological parameter mapping, numerical sensitivity analysis, and physics-informed neural network-based parameter estimation. Methods A coupled fractional Korteweg-de Vries-type system was solved under identical initial and boundary conditions for the three fractional operators. The time-fractional order α was varied over [0.6, 1.0], while the space-fractional order β was varied over [1.5, 2.0]. Simulations used a uniform spatial grid, fixed time step, localized sech 2 initial pulse, and Neumann boundary conditions. The operators were compared using soliton-like velocity, amplitude, pulse width, normalized energy retention, residual error, RMSE, MAE, and CPU runtime. A physics-informed neural network was further used to estimate model parameters while enforcing the fractional PDE residual. Results The Beta derivative produced the most localized and stable soliton-like pulses, with stronger amplitude preservation, lower energy loss, and shorter runtime than the Liouville-Caputo and Atangana-Baleanu formulations under the tested settings. Increasing ephaptic coupling strength reduced pulse amplitude, whereas increasing α improved propagation velocity and increasing β enhanced waveform localization. Quantitative residual and error analyses confirmed that the Beta-based formulation maintained low numerical error while preserving biologically plausible conduction behavior. Conclusion The results support the Beta derivative as a biologically plausible and computationally efficient approximation for soliton-like nerve-pulse propagation in coupled myelinated fibers. The Liouville-Caputo and Atangana-Baleanu operators remain valuable for long-memory and fading-memory regimes, respectively. Future work should integrate literature-constrained biological consistency assessment, stochastic ion-channel dynamics, and heterogeneous multidimensional nerve-bundle geometries.
In medical image analysis, accurate skin lesion categorization is still a major difficulty particularly under limited data conditions and computational complexity. For automated skin cancer detection, in this work we present a scalable hybrid model combining a Convolutional Neural Network (CNN), the Harmonic Mean Optimizer (HMO), and a Support Vector Machine (SVM) classifier—termed HMO-CNN-SVM. Key CNN hyperparameters including learning rate, batch size, and kernel configuration are optimized using the HMO, so greatly boosting classification performance over manual or stationary settings. The model further uses SVM on CNN feature embeddings modified on HMO to improve decision boundary sharpness. Robust performance is shown by experiments carried out on the ACS skin lesion dataset validated by 5-fold cross-valuation and ISIC 2018 benchmarks with an accuracy of 95.02% and consistent generalizing over folds. Crucially, significant parallelism potential made possible by the population-based structure of HMO makes the framework fit for GPU clusters or cloud-based training pipelines. Computational benchmarks expose reasonable overhead in trade for best performance. Thus, the suggested system is a strong contender for implementation in high-performance and distributed computing contexts since it provides both diagnostic dependability and computational tractability.
Anemia remains a critical global health burden, often driven by impaired erythropoietin (EPO) signaling, which reduces red blood cell production. While recombinant EPO therapy is effective, its high cost and associated safety concerns limit its accessibility. This study explores microbial metabolites as affordable and safe alternatives that act as EPO mimetics that can bind and activate the erythropoietin receptor (EPOR). A computational screening of 90 microbial bioactive compounds was conducted, and from those, 16 were selected for detailed analysis. The extracellular domain of EPOR (PDB: 1EBP) was used as the target protein. Molecular docking was performed using AutoDock, followed by ADMET (Absorption, Distribution, Metabolism, Excretion, and Toxicity) profiling with SwissADME and ProTox-III. Protein-protein interaction (PPI) networks were also analyzed in Cytoscape, and the stability of the top complexes was validated via 100 ns molecular dynamics (MD) simulations. Docking results identified Abyssomicin W, Abyssomicin C, and Camptothecin as the top candidates with strong binding affinities (-7.60 kcal/mol) to EPOR. ADMET predictions confirmed their favorable drug-likeness and safety profiles, with Abyssomicin W exhibiting the most promising characteristics, including high gastrointestinal absorption, and no predicted hepatotoxicity or carcinogenicity. PPI network analysis underscored the functional relevance of EPOR in erythropoietic pathways, while molecular dynamics (MD) simulations revealed that Abyssomicin W and Camptothecin formed highly stable complexes with the receptor, whereas the Abyssomicin C complex was unstable. The integrated computational pipeline successfully identified Abyssomicin W as the most stable and promising EPO mimetic candidate. In conclusion, this study identifies Abyssomicin W as a potential and stable EPO mimetic candidate, highlighting the potential of microbial metabolites as cost-effective therapeutics for anemia. Further experimental validation, including direct binding and functional cell-based assays is recommended to confirm its efficacy and safety in biological systems.
Objective This study was aimed to investigate the blood composition and potential mechanism of Liujun Jiaoxian Tang (LJJXT) for treating sepsis. Methods After drug intervention in rats, the main active components in LJJXT liquid and serum were identified by UPLC-QE-MS analysis. The effective components and their targets of LJJXT were further screened through the TCMSP database; the disease-related action targets were retrieved by using the Disgenet and Genecards databases. The intersection of the two sets of targets was taken to construct the “LJJXT-components-targets-diseases” network, PPI diagram, GO and KEGG enrichment analysis diagram. Subsequently, molecular docking studies were conducted on the key targets for treating diseases screened by PPI and the corresponding effective components in LJJXT. Results There were 2159 active ingredients in LJJXT, of which 90 were effective in blood. The 20 screened active ingredients matched 139 targets. There were a total of 2585 disease-related targets, and 76 targets shared by drugs and diseases. There were 2113 biological processes, 43 cell components, 211 molecular functions in GO analysis and 172 pathways obtained by KEGG analysis. The results showed that LJJXT may act on AKT1, TNF, PTGS2 and other targets through the active ingredients in blood such as terpenoids, flavonoids, phenols and alkaloids. It was involved in the regulation of lipid and atherosclerosis, toxoplasmosis, and other signaling pathways to play anti-inflammatory, immune enhancement, reduce oxidative stress and other effects, so as to exert drug efficacy and alleviate sepsis. Molecular docking results showed that kaempferol and vitamin A had high affinity with key therapeutic targets involved in lipid and atherosclerotic signaling pathways, and the combination of kaempferol and JUN was the best. Conclusions This study revealed the effective ingredients and potential mechanisms of LJJXT for treating sepsis, providing sufficient theoretical basis for its clinical treatment of sepsis and subsequent basic research.
Introduction: University students face various stresses, including academic and career anxieties and a lack of interpersonal relationships. These stresses can elevate psychological burdens, negatively affecting their studies and daily lives. Objective: This pilot study aims to quantitatively evaluate the effects of mindful breathing exercises using tablet devices on autonomic nervous system activity in university students by analysis of finger plethysmogram (pulse wave amplitude values) and chaos analysis (Lyapunov exponent and fractal dimension). Methods: In this parallel-group randomized controlled trial, 18 nursing students (Mindful Breathing Group [Mi group], n = 9; control group [nMi group], n = 9) were randomly assigned. On the first day, the Mi group performed mindful breathing, the nMi group performed cross fixation, and finger plethysmogram s were measured. For the next 9 days, the Mi group performed mindful breathing at home before bedtime, while the nMi group performed cross gazing, and finger plethysmograms were measured on days 1 and 9. Data were analyzed using one-way analysis of variance and t -tests. Results: The Mi group showed a significant increase in pulse wave amplitude values over time ( P = .001), whereas the nMi group showed a decrease ( P = .001). Chaos analysis revealed no statistically significant differences between groups in the fractal dimension or Lyapunov exponent. Although descriptive differences were observed, these did not reach statistical significance. Both groups demonstrated positive Lyapunov exponents, suggesting nonlinear characteristics of the pulse wave signals. Conclusions: Mindful breathing using tablet devices may be associated with changes in pulse wave amplitude in university students, which could reflect alterations in peripheral autonomic activity under the present experimental conditions. However, no statistically significant differences were observed in chaos analysis indices. Further research with larger samples and additional physiological measures is required to clarify the relationship between mindful breathing and nonlinear autonomic dynamics. Trial Registration: UMIN Clinical Trials Registry (UMIN000056166; Registered November 15, 2024)
Introduction: Resting-state functional magnetic resonance imaging (rs-fMRI) is widely used to examine functional connectivity (FC) alterations in neurological disorders such as Alzheimer’s disease (AD). Traditional studies either employ whole-brain analyses or focus on specific regions, yet the vast number of FCs and their interrelations complicate interpretation. This study adopts a data-driven, hypothesis-free approach to detect altered functional subnetworks in AD. Methods: Independent component analysis (ICA) was applied to FC matrices from 34 AD patients and 49 healthy controls (HCs) from the Alzheimer’s Disease Neuroimaging Initiative (ADNI). After pruning, significant subnetworks distinguishing AD from HC were identified. Graph theoretical parameters were computed for each subnetwork, and their associations with Mini-Mental State Examination (MMSE) scores were assessed. Results: Three subnetworks effectively differentiated AD patients from HCs. One subnetwork showed significant group differences in network strength, clustering coefficient, and local efficiency, despite no whole-brain differences. Abnormal functional lateralization also emerged within subnetworks. Moreover, FC weights in the identified subnetworks positively correlated with MMSE scores, linking cognitive performance to subnetwork connectivity. Conclusion: These results demonstrate the utility of a data-driven approach in detecting AD-specific altered subnetworks. By providing a modular perspective, this method facilitates targeted examination of connectivity changes, improves interpretability, and deepens understanding of functional disruptions in AD.
Background: Heart disease remains one of the leading causes of mortality worldwide, highlighting the need for early and accurate diagnosis to support effective prevention and treatment strategies. Methods: This study presents a machine-learning-based approach for predicting heart disease using clinical and demographic data from a publicly available dataset. Four widely used classification algorithms—Logistic Regression, Random Forest, K-Nearest Neighbors (KNN), and Decision Trees—were evaluated to identify the most effective predictive model. The dataset underwent comprehensive preprocessing, including handling missing values, categorical encoding, and feature normalization, to enhance data quality and model robustness. Model performance was assessed using accuracy, precision, recall, and AUC-ROC metrics. Results: Findings show that hyperparameter-optimized models, particularly Random Forest and KNN, demonstrated strong predictive performance. Explainability techniques, specifically SHapley Additive exPlanations (SHAP), were incorporated to improve interpretability, transparency, and clinical trust. SHAP values were used to analyze feature importance and provide explanations for individual predictions. Conclusion: The results underscore the potential of interpretable machine-learning models as valuable tools for early diagnosis, risk stratification, and clinical decision support. Future research should employ larger datasets and investigate real-time predictive applications further to enhance the generalizability and clinical utility of these models.
Background: Deep learning has transformed medical imaging by enabling earlier and more accurate disease diagnosis. Lesion and tumor segmentation, essential for analyzing and tracking morphological changes, is commonly done with U-Net variants, though these often lack cross-domain generalization and do not fully leverage foundation models like the Segment Anything Model (SAM), which still requires manual intervention to define the region of interest (ROI). Objectives: To enhance generalization and reduce manual intervention by combining the automatic optimization of nnU-Net with the precision of SAM. Design: Experimental evaluation of a hybrid segmentation framework for lung nodule analysis. Methods: We propose a novel approach integrating the automatic optimization capabilities of nnU-Net for lesion detection with the high-precision segmentation of SAM, eliminating the need for manual intervention by the clinician. The method was evaluated on the LIDC-IDRI dataset, a widely recognized benchmark for lung nodule segmentation. Results: Our approach produces more anatomically coherent segmentations than nnU-Net alone. In many cases, the resulting boundaries more closely reflect true nodule morphology than individual expert annotations, despite high inter-expert variability. Conclusion: The proposed integration of nnU-Net with SAM enables fully automated lesion segmentation without manual intervention. The method improves generalization and accuracy across medical imaging domains, achieving expert-level performance in pulmonary nodule segmentation.
Background: Accurate differentiation between benign and malignant breast tumors is critical for early diagnosis and treatment planning. Traditional approaches often rely on whole-image processing; however, the tumor contour contains rich morphological cues that can independently support malignancy assessment. Leveraging these contour-based features using artificial intelligence (AI) can enhance diagnostic specificity and interpretability. Objective: This study aims to evaluate the diagnostic potential of tumor outline features extracted from mammographic images using deep learning models, with a focus on interpreting their variations numerically and biologically. It also investigates whether combining deep features (ensemble approach) can improve classification accuracy. Methods: A public dataset of 100 mammography tumor contours was analyzed. Eight deep learning models (ResNet50, Xception65, VGG16, AlexNet, DenseNet, GoogLeNet, Inception-v3, and a feature-level ensemble) were used for feature extraction. These features were then classified using 5 machine learning algorithms: SVM, KNN, DT, Naive Bayes, and a shallow neural network. Performance metrics included accuracy, sensitivity, specificity, and precision. Results: Xception65 with Naive Bayes achieved 97.97% accuracy, while the feature ensemble with an ensemble classifier achieved 96.96% accuracy, 95.45% sensitivity, and 98.48% specificity. Naive Bayes consistently outperformed other classifiers in integrating deep contour features. Conclusion and Clinical Interpretation: Tumor contour-based analysis provides biologically meaningful indicators of malignancy—such as irregularity, spiculation, and shape complexity—without relying on full pixel intensity. The results demonstrate that outline-driven AI analysis can enhance breast cancer screening by offering a low-complexity, high-performance diagnostic tool. Future integration into clinical workflows may aid radiologists in real-time and reduce false positives in mammographic diagnosis.
Digital twins (DT) technology has shown considerable growth in recent years. Previous studies have examined technologies in a variety of areas, including health care. However, limited studies have attempted to provide a thorough discussion of strategies for the seamless integration of DT into health care, particularly in the context of interoperability of heterogeneous medical data. This review examines the underlying concept of DT and its possible integration in healthcare, particularly in the context of healthcare interoperability. It also analyzes the main problems such as the lack of standardized protocols, the non-homogeneity of data formats and technical complexity. Finally, potential opportunities are highlighted such as standardized protocol, the creation of an open data platform and the empowerment of semantic interoperability. In conclusion, this review has provided valuable insights for many professionals, including researchers and healthcare providers, which will contribute to empowering patient-centered or personalized medicine and to the development of digital health.
Background:The cancellous tissue forming the inner layer of long bones is highly porous at the center, with porosity decreasing toward the outer layer, leading to gradual variations in mechanical properties. Hence, cancellous tissue can be regarded as a functionally graded material (FGM). This study investigates the mechanical properties of graded cancellous bone. Methods:CT scan images combined with image processing techniques were used to extract gradients in mechanical properties of the femoral neck in bovine samples. Several unit cells were employed to model the microstructure of cancellous bone. The graded properties were validated through both numerical and experimental approaches. Cylindrical models are used for finite element analysis and complementary experimental tests were carried out on the femoral neck region. Results:Analytical relationships for mechanical properties of femur spongy bone have been presented. The Cubic and BCC unit cell structures, with E ave E s I = 187 . 11 and E a v e E s I = 168 . 06 m 4 have maximum and minimum flexural stiffness values, respectively. Also, discrepancies between experimental, analytical, and numerical results were discussed. Conclusions:The tesseract unit cell showed the most similarity with the cancellous bone properties, with only 0.11% difference in flexural stiffness, whereas the cubic unit cell, with an 8.48% difference, was the least suitable for modeling spongy bone.
